College of American Pathologists
Printable Version

  Are we there yet? New respiratory panels


CAP Today




July 2008
Feature Story

William Check, PhD

One wonders what Protagoras would make of the multiplex molecular panels for diagnosing viral respiratory pathogens that are starting to become available. This Greek philosopher is reported to have said, “There are two sides to every question.” Obviously he never asked a group of clinical virologists for their assessment of a complex and challenging new test.

After the half-dozen presentations on multiplex respiratory virus panels at this year’s Clinical Virology Symposium and Pan American Society for Clinical Virology (PASCV) meeting, CAP TODAY asked many knowledgeable laboratorians whether these tests are ready for the clinical virology laboratory. The wide range of opinions expressed by our multiplex panel—which consisted of clinical scientists who have worked with the panels as well as several who are trying to assess the data—would have staggered Protagoras. In fact, any reader who has formed an opinion about these technologies is likely to find at least one expert whose view confirms his or her own diagnosis.

Most current multiplex respiratory virus panels use conventional PCR, detect 15 to 20 pathogens, and perform detection with the Luminex xTAG bead suspension system. Luminex’s xTAG Respiratory Virus Panel (RVP) is the only large panel that is FDA-cleared. EraGen’s MultiCode PLx RVP and Qiagen’s ResPlex II RVP are still for research use only, as is Seegene’s Seeplex Respiratory Pathogen assay, which uses fragment analysis for end detection. Prodesse’s ProFlu+, which is also FDA-cleared, is the only real-time PCR panel; it detects only influenza A/B and respiratory syncytial virus. (Real-time assays are limited by the number of fluorescent dye channels available.) In common understanding “multiplex” describes an assay that detects several analytes while being done in one tube; however, AutoGenomics has a microarray-based assay that detects multiple respiratory viruses on a single chip.

Probably the laboratorian who has the most experience with a multiplex panel is James B. Mahony, PhD, FCCM, FAAM, who has worked extensively with the Luminex xTAG RVP assay, which detects 19 viruses. At PASCV, Dr. Mahony, who is head of service, Regional Virology Laboratory, and professor of pathology and molecular medicine at McMaster University in Hamilton, Ontario, presented data showing that this assay is more sensitive for many respiratory viruses than direct fluorescent antigen (DFA) and culture. “Within five years all labs of all sizes will use multiplex assays for respiratory pathogens,” Dr. Mahony predicts, although he adds, “I don’t think that will happen right away.” In hospitalized children with wheezing and bronchiolitis, he reasons, “the evidence is strong that all pathogens are relevant, so all should be tested for. A clinician cannot tell which virus is infecting the patient based on clinical signs.”

Although RVP was 100 percent sensitive for several of the detected viruses, he says, two viruses with lower sensitivity by the xTAG RVP panel were adenovirus and parainfluenza 3. Dr. Mahony believes that the adenovirus figure was an anomaly based on the comparator test used. He is doing a four-laboratory followup to address adenovirus sensitivity. “My hypothesis is that it will be 90+ percent sensitive like the others,” he says.

Dr. Mahony hopes to be running the Luminex assay routinely by this fall. To justify replacing DFA and culture, he is doing a chart review of more than 800 patients in which the Luminex RVP was used alongside conventional tests. “In our setting it is cost-effective to use multiplex PCR as the frontline test,” he says. He finds a savings in the global hospital budget of several hundred dollars per patient workup.

Julie Fox, BSc, PhD, FRCPath, also reported performance data for the Luminex RVP. “We wanted to challenge the system harder than [DFA and culture],” says Dr. Fox, clinical virologist at the Provincial Laboratory for Public Health in Calgary, Alberta. She developed in-house multiplex real-time PCRs for all viruses in the xTAG panel. Performance of RVP and the in-house PCRs was equivalent for most viruses. One exception was rhinovirus: RVP identified 184 (12 percent) samples as positive for picornaviruses, of which 87 were confirmed as enterovirus or rhinovirus by in-house PCR. RVP missed 14 of 92 RSV-positive specimens, but, Dr. Fox says, those samples had low viral load (low Ct on real-time PCR), so “we were not too concerned about that.” RVP also missed 26 of 62 adenoviruses. These, too, had low viral loads. “We are still running in-house adenovirus PCR alongside the panel,” Dr. Fox adds.

“You don’t lose much sensitivity” with a multiplex platform, she concludes, and it is less cumbersome than maintaining many individual real-time PCR assays.

But she advises thinking carefully about whether that broad testing is needed. For some laboratories a multiplex panel might work as a second-line test to replace culture for samples that are negative on antigen-based tests. Or you could use single molecular assays for common respiratory viruses—flu and respiratory syncytial virus, or RSV—before going to a panel. “Many acute care labs do not need to know about rhino- or corona–viruses,” Dr. Fox says. “Where I work, in a public health lab, we need broader coverage for surveillance. Labs shouldn’t move to [multiplex] technology just because it is the newest, sexiest approach. It brings some of its own issues.” One is longer turnaround time. Dr. Fox’s laboratory sets up the first stage of the multiplex panel in the evening (it is a 24/7 lab) and finishes the next morning, so results are available by midday. “More handling is involved when going away from the real-time format,” she notes.

A U.S. laboratorian who has evaluated Luminex RVP is Christine C. Ginocchio, PhD, director of microbiology, virology, and molecular diagnostics at North Shore-LIJ Health System Laboratories in New York. A study her laboratory conducted demonstrated that the RVP assay detected a significantly higher percentage (68.9 percent) of samples containing a respiratory virus than did R-Mix culture (23.5 percent) and direct immunofluorescence tests (12.2 percent). Dr. Ginocchio has incorporated xTAG RVP into her standard testing algorithm. All patients treated in the emergency room and discharged who are negative on either a rapid flu or RSV test or an eight-virus DFA panel are tested with the RVP panel. All respiratory samples from patients admitted to the hospital and from selected chronic care facilities are also tested with the RVP panel. “RVP is sensitive and specific and provides us with a broad panel of results in 24 hours,” Dr. Ginocchio says.

In her view, whether other laboratories should adopt a multiplex panel depends on their scope of service and patient population. Severity of illness, immune status, HIV infection, and transplant history are important considerations. Pro­desse ProFlu+, the other FDA-cleared multiplex assay, could be appropriate in outpatients, where the laboratory is usually ruling out flu or RSV. It is more limited in scope but provides results in 2.5 to three hours.

“In sicker patients, however, we need a broader scope of detection,” Dr. Ginocchio says. “Adenovirus infection, in particular, can very quickly result in severe outcomes in sicker patients.” Since the detection of adenovirus is critically important, Dr. Ginocchio will participate in the four-site study, coordinated by Dr. Mahony, to further evaluate the sensitivity of the RVP assay for adenovirus detection.

For laboratories without molecular experience, Dr. Ginocchio would recommend starting with ProFlu+, which is simple, faster, and done in a closed system. “Luminex RVP is more complex,” she says, “and could be difficult to start out with.”

A comparison validation study between xTAG RVP and the Qiagen Res­Plex II RVP was described by Melissa B. Miller, PhD, D(ABMM), assistant professor of pathology and laboratory medicine and director of the molecular microbiology laboratory at the University of North Carolina School of Medicine, Chapel Hill. Speaking generally, Dr. Miller says, “The multiplexing technology and power of liquid bead suspension arrays will have a huge impact on testing for respiratory viruses. They diagnose several viruses that we do not detect routinely.” On the other hand, Dr. Miller notes, “It is very clear that not every lab will be doing this methodology.”

She envisions the utility of these assays in patients sick enough to be admitted who appear to have a viral process. “These are largely the patients on whom we are now doing viral cultures,” she says. “Patients whose specimens are rapid antigen-negative and who are admitted can be reflexed to a panel.”

Based on her experience with the two multiplex panels so far, Dr. Miller finds ResPlex II RVP, with six hours from sample to result and about one hour of hands-on time, to be faster and simpler than Luminex xTAG RVP. “On average, it took nine hours to obtain a result with the Luminex RVP,” Dr. Miller says. Also, there is no opening of amplified products with the ResPlex II, reducing risk of contamination, and it differentiates enteroviruses from rhinoviruses. Luminex xTAG RVP, on the other hand, can detect some adenoviruses, it subtypes influenza A H1 from H3, and it is FDA-cleared.

“There is some misperception that these assays are too expensive,” Dr. Miller says. “Clearly they are expensive. But even from the lab perspective they are not more expensive than a combination of methods. And if I would do PCR assays for all of the viral targets in these panels, it would probably cost more than a multiplex kit.”

One issue with a multiplex panel is reporting only the result(s) that the clinician orders or, alternatively, allowing the clinician to order only the whole panel. “Our proposal to the medical staff will be that, if a respiratory virus test gets ordered, every analyte [in the panel] gets ordered,” Dr. Miller says. “That will work out for billing as well.” To convince the physicians of the value of this approach, Dr. Miller plans to share her data on the number of additional pathogens that multiplex panels pick up. “Our physicians are not used to looking routinely for human metapneumovirus or rhino- or enterovirus,” she says. “Once they see the data, I think they would want to know all the results.”

Prodesse’s ProFlu+ multiplex real-time PCR assay is also more sensitive than a combination of DFA and culture, as seen in the three years’ experience Belinda Yen-Lieberman, PhD, has had in offering the ProFlu test in her hospital. Professor of pathology at the Lerner College of Medicine and director of clinical virology, serology, and cellular immunology at the Cleveland Clinic, Dr. Yen-Lieberman says she has been “very happy” with the assay, which entails less hands-on time and decreased risk of laboratory exposure compared with highly multiplexed PCR assays.

She is now going to evaluate the EraGen MultiCode PLx RVP for feasibility and clinical ­utility in the routine clinical laboratory. Dr. Yen-Lieberman says EraGen’s platform looks “very attractive,” being much easier than Luminex RVP, with less washing. She also likes the way the MultiCode PLx software addresses the reporting dilemma. “If a clinician only orders flu or RSV, legally I can’t report the results they didn’t order,” she says. “With EraGen’s software, I can suppress all the unordered results.” Once the specimen is reported as negative for flu and RSV, the physician can order the whole panel as add-on and the rest of the results can be released, reported, and billed.

It is Dr. Yen-Lieberman’s understanding that Era­Gen is submitting the MultiCode PLx RVP for 510(k) clearance during the third quarter of this year.

Systematic evaluation of the performance characteristics of MultiCode PLx RVP was presented at PASCV by Gregory Storch, MD, professor of pediatrics, medicine, and molecular microbiology, director of pediatric infectious diseases, and director of laboratory medicine in pediatrics at Washington University School of Medicine. Dr. Storch used the 17-virus EraGen assay to test a mix of 410 samples evenly divided between positive and negative on an eight-virus DFA screen and culture. For comparison, he developed six multiplex real-time PCRs to cover all 17 viruses. Dr. Storch concluded that EraGen and in-house molecular tests correlated very closely, except that EraGen was better for rhinovirus and in-house PCR was better for parainfluenza. Turnaround time for EraGen was about six hours, plus one to two hours for extraction. “It is pretty much an all-day test,” Dr. Storch says.

He hasn’t implemented any multiplex test yet. “We still haven’t solved workflow and charge issues,” he says. “Ultimately, multiplex platforms will lead to improvements in medical care, but we are in a bit of a tricky period right now.”

In Dr. Storch’s view, multiplex panels “fit very nicely in large reference labs,” which are not trying to achieve faster TAT than once per day. However, he adds, “They still don’t fit in frontline clinical hospital labs, where clinicians are eager to get results very quickly. A test that we set up once a day would result in degradation of our [current] turnaround time. We won’t replace DFA right now.” He sees first-generation multiplex tests as proof of principle: “They are very exciting but not yet mature technology. My crystal ball says improvements will be made to shorten turnaround time.”

Frederick S. Nolte, PhD, D(ABMM), director of clinical laboratories and professor of pathology and laboratory medicine at the Medical University of South Carolina, also evaluated EraGen, running it for two seasons in the background. It increased diagnostic yield, he found, “mostly from better influenza virus detection, plus adding viruses we don’t normally seek with culture,” such as human metapneumovirus, coronavirus, and rhinovirus.

Like Dr. Storch, Dr. Nolte is not planning to implement a multiplex panel now. “One problem with implementing this kind of technology,” he says, “is that it is very technically complex, with a lot of steps and many opportunities for things to go wrong.” In particular, current multiplex assays have many post-amplification manipulations, allowing possible contamination and a potential for false-positive results. “This makes it problematic for all but advanced diagnostic labs,” Dr. Nolte says. Also, analysis time is long relative to real-time PCR standards.

However, Dr. Nolte identifies the major problem as such: “What is the clinical impact of increased diagnostic yield? There is an FDA-cleared assay but without any evidence that it makes a clinical difference.” He wonders whether third-party payers will reimburse for an expensive diagnostic with no demonstrated impact. “Clinical laboratorians believe there is value to [multiplex panels],” he says. “The problem is documenting it. The only study that looked at this used many individual PCR reactions and came to the conclusion that it is not advantageous” (Oosterheert JJ, et al. Clin Inf Dis. 2005;41:1438–1444).

Dr. Nolte acknowledges that respiratory virus detection is moving toward molecular methods, and that the multiplex approach “resonates” because there are so many potential viral respiratory patho­gens. “I remain excited about these tests. But their implementation in clinical labs gives me pause,” he says.

Jacques Corbeil, PhD, Canada Research chair in medical genomics and professor of medicine, Laval University Infectiology Centre, has experience with the AutoGenomics Infiniti RVP multiplex array. For the past two flu seasons he has done beta testing for AutoGenom­ics. He ran samples from 779 symptomatic children under age three who came to the emergency department or to a pediatric clinic. “The assay is highly automated and tests for 25 viruses,” Dr. Corbeil says. It has an eight-hour TAT and processes up to 48 samples per day. Dr. Corbeil found that “it achieved an actual diagnostic in 85 percent of cases.” He created in-house singleplex real-time PCRs for all viruses in the panel. Infiniti RVP and in-house PCRs agreed in 90 percent of cases. Discrepancies usually resulted from singleplex PCRs being more sensitive, with PIV 1–4 especially affected; a lack of sequence information makes it difficult to design primers for these viruses, Dr. Corbeil says. After a third season of testing he hopes to have enough data to demonstrate that rapid diagnostic information reduces pressure to prescribe an antibiotic.

Dr. Corbeil has a Luminex instrument for other applications. “It requires too much hands-on time,” he says. “In a clinical lab you need a lot of automation. The AutoGenomics assay is very automated.”

While several laboratorians designed real-time PCRs for comparison purposes, Richard Hodinka, PhD, uses a suite of molecular tests for routine diagnosis of respiratory viruses. “We have been doing panel respiratory virus testing since long before companies produced such assays,” says Dr. Hodinka, director of the clinical virology laboratory at Children’s Hospital of Philadelphia and associate professor of pediatrics at the University of Pennsylvania School of Medicine. In his current algorithm he first offers a rapid antigen test for RSV. Negative specimens go directly to PCR. Dr. Hodinka offers a panel of real-time PCR assays for RSV A/B, flu A/B, parainfluenza 1–3, adeno­viruses, hMPV, and rhinoviruses. He uses pan-sensitive primers for adenovirus and rhinovirus. “We turn this panel over daily, seven days a week, in multiple batches per day,” Dr. Hodinka says. “Ninety percent of specimens have 18- to 24-hour turnaround time, usually much shorter. This approach has eliminated all culture-based testing for respiratory viruses as well as the vast majority of antigen-based testing in my laboratory.”

However, maintaining a panel of individual molecular assays requires considerable time and effort. “It’s a nonstop job,” Dr. Hodinka says. He agrees with Dr. Fox that maintaining a single multiplex panel is simpler than maintaining many real-time PCRs. On the other hand, he points out, “Look at the large multiplex panels that are now approved or in process. Everyone who has run them is concerned that they are, for the most part, open systems. Mine is a closed system, so I have less of a concern about amplicon contamination.” Some commercial multiplex panels use conventional PCR. “This is where we were many years ago,” Dr. Hodinka says. “It is a step back in time for many of us who now use far more improved technology.”

In Dr. Hodinka’s view, commercial multiplex panels may benefit laboratories who have never done PCR before or have done it in a limited way. “FDA-approved assays have been QC’ed and QA’ed by the company,” he says. “Everything is in place. They should work.” Unfortunately, he says, current multiplex panels are still too complex. “They are not load, press, and go. The technology is not user-friendly enough.”

To Angela Caliendo, MD, PhD, multiplex respiratory virus panels have value for hospitals with transplant and immuno­compro­mised populations and individuals ill enough to be hospitalized. “For healthy adult outpatients the turnaround time is too slow and testing is quite expensive,” says Dr. Caliendo, director of Emory Medical Laboratory and vice chair of pathology and laboratory medicine, Emory University School of Medicine.

Emory’s laboratory now does shell vial culture for respiratory viruses. “This approach has worked well for us,” Dr. Caliendo says. “It is very reliable. But recently we came to appreciate that molecular testing is more sensitive, so we will try to move in that direction in the next year.” Multiplex platforms also provide a broader menu, which may not be needed in all situations. “Physicians who care for bone marrow transplant recipients and hema­tology/on­colo­gy patients are very enthusiastic about this broad panel approach,” Dr. Caliendo reports. “These clinicians are trying to determine whether the res­pira­tory process is infectious or immune mediated. A test that detects additional viruses—human meta­pneumo­virus and corona- and rhino­viruses—could be very useful in this situation. It will help the clini­cian determine whether the patient has an infectious pulmonary problem or some other process that requires different treatment.”

Still, Dr. Caliendo says this of xTAG RVP, the only FDA-cleared highly multiplexed assay: “It is quite labor-intensive. It takes a long time to report out results. And it is not equally sensitive for all viruses.”

Another laboratory director who is reluctant to switch now is Marie Louise Landry, MD, professor and vice chair, Department of Laboratory Medicine, and director of the clinical virology laboratory at Yale University School of Medicine. “Clearly multiplex molecular testing is a direction we need to go in,” Dr. Landry says. “I applaud the companies for tackling the problem. We now have the first generation of their efforts.” Her laboratory will soon evaluate one of the current tests.

But, Dr. Landry says, there are concerns about the tests. “To have a sensitive multiplex PCR that is user-friendly with a short turnaround time would be an advantage. The question is, Are we there yet? There is still a lot of room for improvement.” Her unease about the first generation of multiplex molecular tests is that they use standard PCR, so they take longer, are not as user-friendly, and are more vulnerable to cross-contamination.

Sensitivity is also a worry. Multiplexing creates the potential to lose sensitivity. Published data, she says, support this concern (Li H, et al. J Clin Microbiol. 2007;45:2105–2109). Evaluation of two multiplex tests showed sensitivity for flu A of 83 and 86 percent and for RSV of 63 and 73 percent. “That has me a little concerned,” Dr. Landry says.

Molecular pathologist Karen Kaul, MD, PhD, too, expresses ambivalence about current multi­plex assays for respiratory viruses. “We do need higher-level multiplexing for viral respiratory patho­gens,” says Dr. Kaul, who is board of directors chair of molecular pathology and director of the molecular diagnostics laboratory at Evanston (Ill.) Northwestern Healthcare and professor of pathology and urology at Northwestern University Medical School. “Many patients who present with cough and fever have something viral, but you can’t necessarily predict what you are looking for. Being able to test for many viruses at once is attractive.”

Yet Dr. Kaul lists several downside factors. First, she questions the rigidity of preset panels. “Does it make sense to test for flu in August?” she asks. “Does it make sense to test for SARS right now?” Second, the clinical implications of a positive result are not always clear. “For some viruses our understanding is evolving,” Dr. Kaul says. For hMPV, for instance, there are questions about low-level positives in patients who are not clinically ill, such as family contacts.

Dr. Kaul also worries about losing sensitivity with multiplexing. “Our experience suggests that the ability to detect flu A/B with the Luminex panel is lower than with our in-house real-time PCR assay.” Analytic sensitivity was greater with the in-house assay. “I suspect this is clinically relevant,” Dr. Kaul says, “but it remains to be investigated.”

Like others, Dr. Kaul finds the turnaround time of multiplex platforms unacceptable. “We run flu daily by real-time PCR and try to turn results around in a few hours,” she says. “We tried the Luminex assay on a couple of hundred patients, but we haven’t decided to go live.”

David Hillyard, MD, medical director in the Department of Molecular Pathology at ARUP Laboratories in Salt Lake City, says all labs could potentially benefit from multiplex molecular respiratory virus testing if performance of the assay is adequate. Molecular testing offers greater sensitivity, specificity, and rapidity and allows conventional technology—culture and serology—to be replaced, with potentially much more automatable and standardizable assays. Yet, Dr. Hillyard says, “From a technical perspective, I am not sure that developers have achieved adequate testing formats.”

He is concerned about the complexity of current multiplex assays. “The classic example is the Luminex platform,” Dr. Hillyard says. “It’s a fantastic platform that has a proven track record for immunological testing. For molecular testing, though, it is much more complicated compared to real-time PCR, with much greater demand for hands-on steps. And it takes longer to perform.” Like others, Dr. Hillyard worries about the open platform’s vulnerability to contamination. “It is interesting that in their current formats none of the medium-complexity multiplex assays use the standard chemistry of contamination control, dUTP/UNG enzymology, which has been so important for a whole variety of PCR tests.”

The cost of multiplex testing is another concern. “Traditional testing, especially culture, is also expensive,” he acknowledges. “However, it accommodates more easily the ebb and flow of numbers of clinical samples.” In the middle of respiratory season with a high volume of samples, multiplex panels would probably be cost-effective, having a favorable ratio of samples to controls. “However,” Dr. Hillyard says, “when sample volumes are lower, the cost of respiratory testing may be significantly impacted by the burden of controls on small runs.” One approach is to wait to fill minimum run sizes. But this compromises TAT.

Matthew J. Bankowski, PhD, D(ABMM), says multiplex respiratory virus panels “are still trying to find their place,” and he advises not using them “indiscriminately.” He sees them being used “very selec­tively” in larger hospitals. “Not every patient is at high risk,” says Dr. Bankowski, vice president and technical director of microbiology and infectious disease at Diagnostic Laboratory Services Inc., and in the Department of Pathology at the University of Hawaii, John A. Burns School of Medicine, Honolulu.

“One question on everyone’s mind,” he says, “is what it means when you find coinfection” with the multitarget approach. “Do you treat that patient any differently from finding one of the major pathogens”—flu A/B, RSV, or a PIV? For a transplant patient, having a rhinovirus along with another virus could mean death. On the other hand, he says, “You don’t want to pay $500 for a flu A and a rhinovirus together in a patient who is not severely ill.”

Like many other laboratory directors, Dr. Bankowski is apprehensive about the cost of multiplex panels because of new equipment, training, and labor requirements. The xTAG RVP “appears much more labor-intensive than promoted by the company,” he says. He also sees reimbursement as a big issue. “There is anecdotal evidence that people are getting reimbursement, but also evidence that payers are looking at this very carefully,” he says.

Dr. Bankowski’s reference laboratory has served as a clinical test site for the Prodesse ProFlu+ and hMPV kits and is now a test site for Prodesse’s PIV assay. He notes that an algorithm could be devised in which these tests are performed in serial fashion in the same day using the same extracted nucleic acid.

Nancy Cornish, MD, director of microbiology at Nebraska Method­ist Health System in Omaha, is one of the laboratory directors for whom a multiplex panel is not feasible at this time. “I would love to have multiplex PCR [for respiratory viruses],” says Dr. Cornish. “I think that would answer our problem. But it’s not available in our situation. We don’t have a lot of resources, and we would have to educate our clinicians.”

Dr. Cornish says being able to diagnose a wider array of respiratory viruses would improve patient care. Her ICU physicians have asked about it. However, she is uneasy about volume. “Luminex is a good example,” she says. “We have a doctoral-level scientist who specializes in molecular, so I would feel comfortable bringing that test in-house. But what else would we run on it? We couldn’t just do respiratory virus panels in season.”

And what about charging for the tests? “How do you charge the patient if the clinician didn’t order all 20 viruses?” she asks. Some physicians say they don’t want to know about all viruses. “That’s a problem for me,” Dr. Cornish says. “If I know a virus not ordered comes up positive, I feel obligated to call the doctor. Then the doctor can order the panel. But do we have a right to charge the patient for 20 viruses?”

A laboratorian at a community hospital who has instituted a multiplex panel is Holly Alexander, PhD, clinical scientist at Via Christi Regional Medical Center, Wichita, Kan. “A couple of years ago when I first got interested in multiplex, FDA clearance was an issue,” she says. FDA clearance makes it easier to do validation and verification. Dr. Alexander started working with the xTAG RVP and was ready to go live when it became FDA-cleared early this year. She didn’t have time to train people until flu season slacked off. “We started picking up a lot of human meta­pneumovirus and rhinovirus that we had not seen in culture,” she says. “That’s very exciting.” Dr. Alexander finds that half of DFA-negative specimens have something in them. ”You really feel that you have improved your contribution to the care of patients,” she says.

Dr. Alexander has eliminated shell vial and tube culture. “So we have eliminated a lot of work and made it easier to train techs,” she says. “It takes a lot of time to develop the expertise to read CPE in a virus tube.” Most samples are now coming from one 500-bed hospital. “If a hospital is big enough that the lab does [complete] virus cultures, it probably is big enough to use this test,” Dr. Alexander says.

Technically, she calls this procedure “a bear.” It requires “great attention to detail,” she says, particularly in avoiding contamination. “But it pays off in getting results at the end.”

Many newer technologies are under development. Dr. Cornish reports that Cepheid plans to have 20 assays for the GeneXpert by 2012, including some for respiratory viruses. Dr. Hillyard likes upcoming technology based on capillary electrophoresis because “it has great bandwidth and allows you to see many amplicons in a single interrogation,” he says. Detection with mass spectrometry is also under study. “Ibis Biosciences has worked with highly multiplexed PCR and detection with electrospray ionization mass spectrometry,” Dr. Hillyard says. “Initial results with influenza look very promising.” He also thinks Idaho Technology’s “assay in a pouch” is promising, especially for near-patient work.

In his view, “Because multiplex testing is so important to many areas of pathology, including oncology and human genetics, improvements in technology will spill over to give us better molecular viral respiratory tests.” However, he adds, “To my mind we still need a kind of a sea change in technology that can embrace all of the difficulties that [current methods face] and allow for multiplex testing that will be cost-effective and convenient to labs.”

Of the current multiplex platforms for respiratory viruses, Dr. Alexander says, “This is the forefront of a lot of exciting new testing that is coming. It’s a good time to be in virology.”

William Check is a medical writer in Wilmette, Ill.